1. Hardware Lesson CS1313 Spring 2015 1 Hardware Outline 1.Hardware Outline 2.What is a Computer? 3.Components of a Computer 4.Categories of Computer Hardware 5.Central Processing Unit (CPU) 6.CPU Examples 7.CPU Parts 8.CPU: Control Unit 9.CPU: Arithmetic/Logic Unit 10.CPU: Registers 11.How Registers Are Used 12.Multicore 13.Multicore History 14.Storage 15.Primary Storage 16.Cache 17.From Cache to the CPU 18.Main Memory (RAM) 19.Main Memory Layout 20.RAM vs ROM 21.Speed => Price => Size 22.How Data Travel Between RAM and CPU 23.Loading Data from RAM into the CPU 24.RAM is Slow 25.Why Have Cache? 26.Secondary Storage 27.Media Types 28.Speed, Price, Size 29.CD-ROM & DVD-ROM 30.CD-ROM & DVD-ROM: Disadvantage 31.CD-ROM & DVD-ROM: Advantages 32.Why Are Floppies So Expensive Per MB? 33.I/O 34.I/O: Input Devices 35.I/O: Output Devices 36.Bits 37.Bytes 38.Words 39.Putting Bits Together 40.Putting Bits Together (cont’d) 41.Powers of 2 42.Powers of 2 vs Powers of 10 43.KB, MB, GB, TB, PB 44.Kilo, Mega, Giga, Tera, Peta 45.EB, ZB, YB 46.Moore’s Law 47.Implication of Moore’s Law 48.Double, double, …

2. Hardware Lesson CS1313 Spring 2015 2 What is a Computer? “… [A] programmable electronic device that can store, retrieve and process data.” (N. Dale & D. Orshalick, Introduction to PASCAL and Structured Design, D.C. Heath & Co., Lexington MA, 1983, p. 2)

3. Hardware Lesson CS1313 Spring 2015 3 Components of a Computer Computer Hardware Physical Devices Software Instructions & Data DON’T PANIC! This discussion may be confusing at the moment; it’ll make more sense after you’ve written a few programs.

4. Hardware Lesson CS1313 Spring 2015 4 Categories of Computer Hardware Central Processing Unit (CPU) Storage Primary: Cache, RAM Secondary: Hard disk, removable (e.g., CD) I/O Input Devices Output Devices

5. Hardware Lesson CS1313 Spring 2015 5 Central Processing Unit (CPU) The Central Processing Unit (CPU), also called the processor, is the “brain” of the computer. Intel Haswell quad core innards http://regmedia.co.uk/2013/05/23/haswell_1.jpg Haswell exterior http://www9.pcmag.com/media/images/387706-intel-haswell.jpg?thumb=y

6. Hardware Lesson CS1313 Spring 2015 6 CPU Examples x86: Intel Celeron/Pentium/Core/Atom/i3/i5/i7/Xeon and AMD Athlon/Sempron/Turion/Phenom/Opteron (and related “x86” models from smaller manufacturers) (Windows, MacOS and Linux PCs; some Android tablets) http://en.wikipedia.org/wiki/X86 ARM (in 95% of smartphones, plus many tablets) http://en.wikipedia.org/wiki/ARM_processor IBM POWER8 (servers) https://en.wikipedia.org/wiki/Power_Architecture ia64: Intel Itanium (servers) http://en.wikipedia.org/wiki/Itanium Oracle SPARC64 X+ (servers) http://en.wikipedia.org/wiki/Sparc

7. Hardware Lesson CS1313 Spring 2015 7 CPU Parts Arithmetic/Logic UnitControl Unit Registers Fetch Next Instruction AddSub MultDiv AndOr Not … Integer Floating Point Fetch DataStore Data Increment Instruction Ptr Execute Instruction … The CPU consists of three main parts: Control Unit Arithmetic/Logic Unit Registers

8. Hardware Lesson CS1313 Spring 2015 8 CPU: Control Unit The Control Unit decides what to do next. For example: memory operations: for example, load data from main memory (RAM) into the registers; store data from the registers into main memory; arithmetic/logical operations: e.g., add, multiply; branch: choose among several possible courses of action.

9. Hardware Lesson CS1313 Spring 2015 9 CPU: Arithmetic/Logic Unit The Arithmetic/Logic Unit (ALU) performs arithmetic and logical operations. Arithmetic operations: e.g., add, subtract, multiply, divide, square root, cosine, etc. Logical operations: e.g., compare two numbers to see which is greater, check whether a true/false statement is true, etc.

10. Hardware Lesson CS1313 Spring 2015 10 CPU: Registers Registers are memory-like locations inside the CPU where data and instructions reside that are being used right now. That is, registers hold the operands being used by the current arithmetic or logical operation, or the result of the arithmetic or logical operation that was just performed. For example, if the CPU is adding two numbers, then the addend is in some register; the augend is in another register; after the addition is performed, the sum shows up in yet another register. A typical CPU has only a few hundred to a few thousand bytes of registers.

11. Hardware Lesson CS1313 Spring 2015 11 How Registers Are Used Every arithmetic or logical operation has one or more operands and one result. Operands are contained in registers (“source”). A “black box” of circuits performs the operation. The result goes into a register (“destination”). Example: addend in R0 augend in R1 ADD sum in R2 5 7 12 Register Ri Register Rj Register Rk operand result Operation circuitry

12. Hardware Lesson CS1313 Spring 2015 12 Multicore A multicore CPU is a chip with multiple, independent “brains,” known as cores. These multiple cores can run completely separate programs, or they can cooperate together to work simultaneously in parallel on different parts of the same program. All of the cores share the same connection to memory – and the same bandwidth (memory speed).

13. Hardware Lesson CS1313 Spring 2015 13 Multicore History (x86) Single core: November 1971 (Intel 4004) Dual core: October 2005 (Intel), March 2006 (AMD) Quad core: June 2006 (Intel), Sep 2007 (AMD) Hex core: Sep 2008 (Intel), June 2009 (AMD) 8 core (Intel & AMD), 12 core (AMD only): March 2010 16 core: Nov 2011 (AMD only) 18 core: Sep 2014 (Intel only) http://www.intel.com/pressroom/kits/quickreffam.htmhttp://www.intel.com/pressroom/kits/quickreffam.htm (dual core, quad core) http://ark.intel.com/products/family/34348/Intel-Xeon-Processor-7000-Sequence#@Serverhttp://ark.intel.com/products/family/34348/Intel-Xeon-Processor-7000-Sequence#@Server (6 core) http://ark.intel.com/ProductCollection.aspx?familyID=594&MarketSegment=SRV http://ark.intel.com/ProductCollection.aspx?familyID=594&MarketSegment=SRV (8 core) http://en.wikipedia.org/wiki/Intel_Nehalem_(microarchitecture)http://en.wikipedia.org/wiki/Intel_Nehalem_(microarchitecture) (8 core) http://en.wikipedia.org/wiki/AMD_Opteronhttp://en.wikipedia.org/wiki/AMD_Opteron (12 core) Note that this is only for “x86” – other processor families (for example, POWER) introduced multicore earlier.

14. Hardware Lesson CS1313 Spring 2015 14 Storage There are two major categories of storage: Primary Cache Main memory (RAM) Secondary Hard disk Removable (e.g., CD, floppy)

15. Hardware Lesson CS1313 Spring 2015 15 Primary Storage Primary storage is where data and instructions reside when they’re being used by a program that is currently running. Typically is volatile: The data disappear when the power is turned off. Typically comes in two subcategories: Cache Main memory (RAM)

16. Hardware Lesson CS1313 Spring 2015 16 Cache Cache memory is where data and instructions reside when they are going to be used very very soon, or have just been used. Cache is very fast (typically 5% - 100% of the speed of the registers) compared to RAM (~1% of the speed of the registers). Therefore, it’s very expensive (e.g., $38 per MB) http://ark.intel.com/products/77491/Intel-Core-i3-4350-Processor-4M-Cache-3_60-GHzhttp://ark.intel.com/products/77491/Intel-Core-i3-4350-Processor-4M-Cache-3_60-GHz ($149.00) http://ark.intel.com/products/77488/Intel-Core-i3-4160-Processor-3M-Cache-3_60-GHzhttp://ark.intel.com/products/77488/Intel-Core-i3-4160-Processor-3M-Cache-3_60-GHz ($134.79) http://www.pricewatch.com/ Therefore, it’s very small (e.g., 1 / 4 MB to 64 MB) … but still much bigger than registers.

17. Hardware Lesson CS1313 Spring 2015 17 From Cache to the CPU Typically, data move between cache and the CPU at speeds closer to that of the CPU performing calculations. CPU Cache 46 GB/sec (7%) on a 1.7 GHz i3 Haswell Up to 653 GB/sec on a 1.7 GHz Pentium B940 http://content.hwigroup.net/images /products/xl/204419/dell_latitude_ e5540_55405115.jpg

18. Hardware Lesson CS1313 Spring 2015 18 Main Memory (RAM) Main memory (RAM) is where data and instructions reside when a program that is currently running is going to use them at some point during the run (whether soon or not). Much slower than cache (e.g., less than 1% of CPU speed for RAM, vs 5-100% of CPU speed for cache) Therefore, much cheaper than cache (e.g., $0.0084/MB for RAM vs $38/MB for cache) http://www.pricewatch.com/, http://www.ebay.com/ http://www.pricewatch.com/http://www.ebay.com/ Therefore, much larger than cache (e.g., one to hundreds of GB for RAM vs 1 / 4 MB to 64 MB for cache)

19. Hardware Lesson CS1313 Spring 2015 19 Main Memory Layout Main memory is made up of locations, also known as cells. Each location has a unique integer address that never changes. Each location has a value – also known as the contents – that the CPU can look at and change. We can think of memory as one contiguous line of cells.

20. Hardware Lesson CS1313 Spring 2015 20 RAM vs ROM RAM: Random Access Memory Memory that the CPU can look at and change arbitrarily (i.e., can load from or store into any location at any time, not just in a sequence). We often use the phrases Main Memory, Memory and RAM interchangeably. Sometimes known as core memory, named for an older memory technology. (Note that this use of the word “core” is unrelated to “dual core.”) ROM: Read Only Memory Memory that the CPU can look at arbitrarily, but cannot change.

21. Hardware Lesson CS1313 Spring 2015 21 Speed => Price => Size Registers are VERY fast, because they are etched directly into the CPU. Cache is also very fast, because it’s also etched into the CPU, but it isn’t directly connected to the Control Unit or Arithmetic/Logic Unit. Cache operates at speeds similar to registers, but cache is MUCH bigger than the collection of registers (typically on the order of 1,000 to 10,000 times as big). Main memory (RAM) is much slower than cache, because it isn’t part of the CPU; therefore, it’s much cheaper than cache, and therefore it’s much bigger than cache (for example, 1,000 times as big).

22. Hardware Lesson CS1313 Spring 2015 22 How Data Travel Between RAM and CPU CPU The bus is the connection from the CPU to main memory; all data travel along it. For now, we can think of the bus as a big wire connecting them.

23. Hardware Lesson CS1313 Spring 2015 23 Loading Data from RAM into the CPU

24. Hardware Lesson CS1313 Spring 2015 24 RAM is Slow CPU 15 GB/sec (2.3%) Up to 653 GB/sec on a 1.7 GHz i3 Haswell Richard Gerber, The Software Optimization Cookbook: High-performance Recipes for the Intel Architecture. Intel Press, 2002, pp. 161-168. Bottleneck The speed of data transfer between Main Memory and the CPU is much slower than the speed of calculating, so the CPU spends most of its time waiting for data to come in or go out.

25. Hardware Lesson CS1313 Spring 2015 25 Why Have Cache? CPU Cache is much faster than RAM, so the CPU doesn’t have to wait nearly as long for stuff that’s already in cache: it can do more operations per second! 46 GB/sec (7%) Richard Gerber, The Software Optimization Cookbook: High-performance Recipes for the Intel Architecture. Intel Press, 2002, pp. 161-168. Up to 653 GB/sec on a 1.7 GHz i3 Haswell 15 GB/sec (2.3%)

26. Hardware Lesson CS1313 Spring 2015 26 Secondary Storage Where data and instructions reside that are going to be used in the future Nonvolatile: data don’t disappear when power is turned off. Much slower than RAM, therefore much cheaper, therefore much larger. Other than hard disk, most are portable: they can be easily removed from your computer and taken to someone else’s.

27. Hardware Lesson CS1313 Spring 2015 27 Media Types Solid State Always can be read Always can be written and rewritten multiple times Contents don’t degrade much over time Can’t be erased by magnets Magnetic Always can be read Always can be written and rewritten multiple times Contents degrade relatively rapidly over time Can be erased by magnets Optical Always can be read Some can be written only once, some can be rewritten multiple times Contents degrade more slowly than magnetic media Can’t be erased by magnets Paper: forget about it!

28. Hardware Lesson CS1313 Spring 2015 28 Speed, Price, Size Medium Speed (MB/sec) Size (MB)Media Type Can write to it? Port- able? Pop- ular? Drive cost ($) Media cost ($/MB) Cache46,00032L1/L2/L3YNReq’d$38 RAM15,000786,432DDR4YNReq’d$0.0084 USB 3 Flash625512,000SolidYYY$0.0003 Hard Disk1006,000,000MagYNReq’d$0.00003 Blu-ray72100,000OptYYN$55$0.000036 DVD+RW3217,000OptYYY$10$0.000045 CD-RW7.8700OptYYY$10$0.00033 Mag tape1602,500,000MagYYN$906$0.00002 Floppy0.031.44MagYYY$5$0.24 Cassette<< 1 MagYYHistorical Paper tape<< 1 PaperYYHistorical Punch card<< 1 PaperYYHistorical * Maximum among models commonly available for PCs Note: All numbers are approximate as of Jan 2015 (amazon.com, bestbuy.com, cendyne.com, creativelabs.com, dell.com, pcworld.com, pricewatch.com, rakuten.com, sony.com, storagetek.com, toshiba.com, wikipedia.org). Tape drive and cartridge are LTO-5.

29. Hardware Lesson CS1313 Spring 2015 29 CD-ROM/DVD-ROM/BD-ROM When a CD or DVD or Blu-ray holds data instead of music or a movie, it acts very much like Read Only Memory (ROM): it can only be read from, but not written to; it’s nonvolatile; it can be addressed essentially arbitrarily (it’s not actually arbitrary, but it’s fast enough that it might as well be).

30. Hardware Lesson CS1313 Spring 2015 30 CD-ROM/DVD-ROM/BD-ROM: Disadvantage Disadvantage of CD-ROM/DVD-ROM/BD-ROM compared to ROM: Speed: CD-ROM/DVD-ROM/BD-ROM are much slower than ROM: CD-ROM is 7.6 MB/sec (peak); DVD-ROM is 16 MB/sec; BD- ROM is 17 MB/sec. Most ROM these days is 20-80 MB/sec (1000+ times as fast as DVD or Blu-ray and 2000+ times as fast as CD).

31. Hardware Lesson CS1313 Spring 2015 31 CD-ROM & DVD-ROM: Advantages Advantages of CD-ROM/DVD-ROM compared to ROM: Price: CD-ROM and DVD-ROM are much cheaper than ROM. Blank DVDs and blank BDs are roughly $0.00004 per MB; blank CDs are roughly $0.0003 per MB. ROM is even more expensive than RAM (which is $0.0084/MB), because it has to be made special. Size: CD-ROM and DVD-ROM are much larger – they can have arbitrary amount of storage (on many CDs or DVDs); ROM is limited to a few GB.

32. Hardware Lesson CS1313 Spring 2015 32 Why Are Floppies So Expensive Per MB? DVDs cost roughly $0.00005 per MB, but floppy disks cost about $0.24 per MB, over 16,000 times as expensive per MB. Why? Well, an individual DVD has much greater capacity than an individual floppy (4.7-17 GB vs. 1.44 MB), and the costs of manufacturing the actual physical objects are similar. And, because floppies are much less popular than CDs, they aren’t manufactured in high quantities – so it’s tricky to amortize the high fixed costs of running the factory. So, the cost of a floppy per MB is much higher.

33. Hardware Lesson CS1313 Spring 2015 33 I/O We often say I/O as a shorthand for “Input/Output.”

34. Hardware Lesson CS1313 Spring 2015 34 I/O: Input Devices We often say I/O as a shorthand for “Input/Output.” Input Devices transfer data into computer (e.g., from a user into memory). For example: Keyboard Mouse Scanner Microphone Touchpad Joystick

35. Hardware Lesson CS1313 Spring 2015 35 I/O: Output Devices We often say I/O as a shorthand for “Input/Output.” Output Devices transfer data out of computer (e.g., from memory to a user). For example: Monitor Printer Speakers NOTE: A device can be both input and output – for example, a touchscreen.

36. Hardware Lesson CS1313 Spring 2015 36 Bits Bit (Binary digIT) Tiniest possible piece of memory. Made of teeny tiny transistors wired together. Has 2 possible values that we can think of in several ways: Low or High: Voltage into transistor Off or On: Conceptual description of transistor state False or True: Boolean value for symbolic logic 0 or 1: Integer value Bits aren’t individually addressable: the CPU can’t load from or store into an individual bit of memory.

37. Hardware Lesson CS1313 Spring 2015 37 Bytes Byte: a sequence of 8 contiguous bits (typically) On most platforms (kinds of computers), it’s the smallest addressable piece of memory: typically, the CPU can load from or store into an individual byte. Possible integer values: 0..255 or -128..127 (to be explained later) Can also represent a character (e.g., letter, digit, punctuation; to be explained later)

38. Hardware Lesson CS1313 Spring 2015 38 Words Word: a sequence of 4 or 8 contiguous bytes (typically); that is, 32 or 64 contiguous bits Standard size for storing a number (integer or real) Standard size for storing an address (special kind of integer)

39. Hardware Lesson CS1313 Spring 2015 39 Putting Bits Together 1 bit: 2 1 = 2 possible values: or 2 bits: 2 2 = 4 possible values 3 bits: 2 3 = 8 possible values 01 00001111 000001011001 100101110111

40. Hardware Lesson CS1313 Spring 2015 40 Putting Bits Together (cont’d) 4 bits: 2 4 = 16 possible values … 8 bits: 2 8 = 256 possible values … 10 bits: 2 10 = 1,024 possible values … 16 bits: 2 16 = 65,536 possible values … 32 bits: 2 32 = 4,294,967,296 possible values (typical size of an integer in most computers today)

41. Hardware Lesson CS1313 Spring 2015 41 Powers of 2 2020 =12 11 =2,048 2121 =22 12 =4,096 2 =42 13 =8,192 2323 =82 14 =16,384 2424 =162 15 =32,768 2525 =322 16 =65,536 2626 =642 17 =131,072 2727 =1282 18 =262,144 2828 =2562 19 =524,288 2929 =5122 20 =1,048,576 2 10 =1,024(about a thousand) (about a million)

42. Hardware Lesson CS1313 Spring 2015 42 Powers of 2 vs Powers of 10 A rule of thumb for comparing powers of 2 to powers of 10: 2 10 ~ 10 3 So: 2 10 ~ 1,000 (thousand) 2 20 ~ 1,000,000 (million) 2 30 ~ 1,000,000,000 (billion) 2 40 ~ 1,000,000,000,000 (trillion) 2 50 ~ 1,000,000,000,000,000 (quadrillion) 2 60 ~ 1,000,000,000,000,000,000 (quintillion)

43. Hardware Lesson CS1313 Spring 2015 43 KB, MB, GB, TB, PB Kilobyte (KB): 2 10 bytes, which is approximately 1,000 bytes (thousand) Megabyte (MB): 2 20 bytes, which is approximately 1,000,000 bytes (million) Gigabyte (GB): 2 30 bytes, which is approximately 1,000,000,000 bytes (billion) Terabyte (TB): 2 40 bytes, which is approximately 1,000,000,000,000 bytes (trillion) Petabyte (PB): 2 50 bytes, which is approximately 1,000,000,000,000,000 bytes (quadrillion)

44. Hardware Lesson CS1313 Spring 2015 44 Kilo, Mega, Giga, Tera, Peta Kilobyte (KB):2 10 bytes =1,024 bytes ~1,000 bytes Approximate size: one e-mail (plain text) Desktop Example: TRS-80 w/4 KB RAM (1977) Megabyte (MB):2 20 bytes =1,048,576 bytes ~1,000,000 bytes Approximate size: 30 phonebook pages Desktop Example: IBM PS/2 PC w/1 MB RAM (1987) Gigabyte (GB):2 30 bytes =1,073,741,824 bytes ~1,000,000,000 bytes Approximate size: 15 copies of the OKC white pages Desktop: c. 1997 Terabyte (TB):2 40 bytes =1,099,511,627,776 bytes ~1,000,000,000,000 bytes Approximate size: 5,500 copies of a phonebook listing everyone in the world Desktop: Example: IBM x3500M4 (Jan 2013: 768 GB) Petabyte (PB):2 50 bytes ~1,000,000,000,000,000 bytes Desktop: ???

45. EB, ZB, YB Exabyte (EB): 2 60 bytes, which is approximately 1,000,000,000,000,000,000 bytes (quintillion) (global data volume in 2010; global daily Internet traffic was ~1.7 EB in 2013; ~20,000 copies of every book ever written) Zettabyte (ZB): 2 70 bytes, which is approximately 1,000,000,000,000,000,000,000 bytes (sextillion) (By late 2017, annual Internet traffic will be ~1 ZB.) Yottabyte (YB): 2 80 bytes, which is approximately 1,000,000,000,000,000,000,000,000 bytes (septillion) (At current growth rates, by 2047, annual Internet traffic will be ~1 YB.) http://en.wikipedia.org/wiki/Exabyte http://www.cisco.com/web/solutions/sp/vni/vni_forecast_highlights/ Hardware Lesson CS1313 Spring 2015 45

46. Hardware Lesson CS1313 Spring 2015 46 Moore’s Law Moore’s Law: Computing speed and capacity double every 24 months. In 1965 Gordon Moore (Chairman Emeritus, Intel Corp) observed the “doubling of transistor density on a manufactured die every year.” People have noticed that computing speed and capacity are roughly proportional to transistor density. Moore’s Law is usually hedged by saying that computing speed doubles every 18-24 months (typically 24). See: http://www.intel.com/content/www/us/en/silicon- innovations/moores-law-technology.html http://www.intel.com/pressroom/kits/quickreffam.htm http://en.wikipedia.org/wiki/Transistor_count http://en.wikipedia.org/wiki/Beckton_%28microprocessor%29#6500.2F7500-series_.22Beckton.22

47. Hardware Lesson CS1313 Spring 2015 47 Implication of Moore’s Law If computing speed and capacity double every 24 months, what are the implications in our lives? Well, the average undergrad student is – to one significant figure – about 20 years old. And the average lifespan in the US – to one significant figure – is about 80 years. So, the average undergrad student has 60 years to go. So how much will computing speed and capacity increase during the time you have left?

48. Hardware Lesson CS1313 Spring 2015 48 Double, double, … 60 years / 2 years = 30 doublings What is 2 30 ? Consider the computer on your desktop today, compared to the computer on your desktop the day you die. How much faster will it be? Can we possibly predict what the future of computing will enable us to do?